Total Complexity | 112 |
Total Lines | 656 |
Duplicated Lines | 0 % |
Changes | 0 |
Complex classes like Lexer often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use Lexer, and based on these observations, apply Extract Interface, too.
1 | <?php |
||
8 | class Lexer implements LexerInterface |
||
9 | { |
||
10 | /** |
||
11 | * TODO: Document this |
||
12 | * |
||
13 | * @var Source|null |
||
14 | */ |
||
15 | protected $source; |
||
16 | |||
17 | /** |
||
18 | * TODO: Document this |
||
19 | * |
||
20 | * @var string|null |
||
21 | */ |
||
22 | protected $body; |
||
23 | |||
24 | /** |
||
25 | * TODO: Document this |
||
26 | * |
||
27 | * @var int |
||
28 | */ |
||
29 | protected $bodyLength; |
||
30 | |||
31 | /** |
||
32 | * @var array |
||
33 | */ |
||
34 | protected $options = []; |
||
35 | |||
36 | /** |
||
37 | * The previously focused non-ignored token. |
||
38 | * |
||
39 | * @var Token |
||
40 | */ |
||
41 | protected $lastToken; |
||
42 | |||
43 | /** |
||
44 | * The currently focused non-ignored token. |
||
45 | * |
||
46 | * @var Token |
||
47 | */ |
||
48 | protected $token; |
||
49 | |||
50 | /** |
||
51 | * TODO: Document this |
||
52 | * |
||
53 | * @var int |
||
54 | */ |
||
55 | protected $pos; |
||
56 | |||
57 | /** |
||
58 | * The (1-indexed) line containing the current token. |
||
59 | * |
||
60 | * @var int |
||
61 | */ |
||
62 | protected $line; |
||
63 | |||
64 | /** |
||
65 | * The character offset at which the current line begins. |
||
66 | * |
||
67 | * @var int |
||
68 | */ |
||
69 | protected $lineStart; |
||
70 | |||
71 | /** |
||
72 | * Lexer constructor. |
||
73 | * |
||
74 | * @param TokenReaderInterface $reader |
||
75 | */ |
||
76 | public function __construct() |
||
77 | { |
||
78 | $startOfFileToken = new Token(TokenKindEnum::SOF); |
||
79 | |||
80 | $this->lastToken = $startOfFileToken; |
||
81 | $this->token = $startOfFileToken; |
||
82 | $this->line = 1; |
||
83 | $this->lineStart = 0; |
||
84 | } |
||
85 | |||
86 | /** |
||
87 | * @inheritdoc |
||
88 | * @throws SyntaxErrorException |
||
89 | */ |
||
90 | public function advance(): Token |
||
91 | { |
||
92 | $this->lastToken = $this->token; |
||
93 | return $this->token = $this->lookahead(); |
||
94 | } |
||
95 | |||
96 | /** |
||
97 | * @inheritdoc |
||
98 | * @throws SyntaxErrorException |
||
99 | */ |
||
100 | public function lookahead(): Token |
||
101 | { |
||
102 | $token = $this->token; |
||
103 | |||
104 | if (TokenKindEnum::EOF !== $token->getKind()) { |
||
105 | do { |
||
106 | $next = $this->readToken($token); |
||
107 | $token->setNext($next); |
||
108 | $token = $next; |
||
109 | } while (TokenKindEnum::COMMENT === $token->getKind()); |
||
110 | } |
||
111 | |||
112 | return $token; |
||
113 | } |
||
114 | |||
115 | /** |
||
116 | * @inheritdoc |
||
117 | */ |
||
118 | public function getOption(string $name, $default = null) |
||
119 | { |
||
120 | return $this->options[$name] ?? $default; |
||
121 | } |
||
122 | |||
123 | /** |
||
124 | * @inheritdoc |
||
125 | * @throws LanguageException |
||
126 | */ |
||
127 | public function getBody(): string |
||
128 | { |
||
129 | return $this->getSource()->getBody(); |
||
130 | } |
||
131 | |||
132 | /** |
||
133 | * @inheritdoc |
||
134 | */ |
||
135 | public function getTokenKind(): string |
||
136 | { |
||
137 | return $this->token->getKind(); |
||
138 | } |
||
139 | |||
140 | /** |
||
141 | * @inheritdoc |
||
142 | */ |
||
143 | public function getTokenValue(): ?string |
||
144 | { |
||
145 | return $this->token->getValue(); |
||
146 | } |
||
147 | |||
148 | /** |
||
149 | * @inheritdoc |
||
150 | */ |
||
151 | public function getToken(): Token |
||
152 | { |
||
153 | return $this->token; |
||
154 | } |
||
155 | |||
156 | /** |
||
157 | * @inheritdoc |
||
158 | * @throws LanguageException |
||
159 | */ |
||
160 | public function getSource(): Source |
||
161 | { |
||
162 | if ($this->source instanceof Source) { |
||
163 | return $this->source; |
||
164 | } |
||
165 | |||
166 | throw new LanguageException('No source has been set.'); |
||
167 | } |
||
168 | |||
169 | /** |
||
170 | * @inheritdoc |
||
171 | */ |
||
172 | public function getLastToken(): Token |
||
173 | { |
||
174 | return $this->lastToken; |
||
175 | } |
||
176 | |||
177 | /** |
||
178 | * @inheritdoc |
||
179 | */ |
||
180 | public function setSource(Source $source) |
||
181 | { |
||
182 | $this->body = $source->getBody(); |
||
183 | $this->bodyLength = \strlen($this->body); |
||
184 | $this->source = $source; |
||
185 | return $this; |
||
186 | } |
||
187 | |||
188 | /** |
||
189 | * @inheritdoc |
||
190 | */ |
||
191 | public function setOptions(array $options) |
||
195 | } |
||
196 | |||
197 | /** |
||
198 | * @param Token $prev |
||
199 | * @return Token |
||
200 | * @throws SyntaxErrorException |
||
201 | */ |
||
202 | protected function readToken(Token $prev): Token |
||
203 | { |
||
204 | $this->pos = $prev->getEnd(); |
||
205 | |||
206 | $this->skipWhitespace(); |
||
207 | |||
208 | $line = $this->line; |
||
209 | $col = 1 + $this->pos - $this->lineStart; |
||
210 | |||
211 | if ($this->pos >= $this->bodyLength) { |
||
212 | return $this->createEndOfFileToken($line, $col, $prev); |
||
213 | } |
||
214 | |||
215 | $code = charCodeAt($this->body, $this->pos); |
||
|
|||
216 | |||
217 | // Punctuation: [!$&:=@|()[]{}]{1} |
||
218 | if (33 === $code || 36 === $code || 38 === $code || 58 === $code || 61 === $code || 64 === $code || 124 === $code || |
||
219 | 40 === $code || 41 === $code || 91 === $code || 93 === $code || 123 === $code || 125 === $code) { |
||
220 | return $this->lexPunctuation($code, $line, $col, $prev); |
||
221 | } |
||
222 | |||
223 | // Comment: #[\u0009\u0020-\uFFFF]* |
||
224 | if (35 === $code) { |
||
225 | return $this->lexComment($line, $col, $prev); |
||
226 | } |
||
227 | |||
228 | // Int: -?(0|[1-9][0-9]*) |
||
229 | // Float: -?(0|[1-9][0-9]*)(\.[0-9]+)?((E|e)(+|-)?[0-9]+)? |
||
230 | if ($code === 45 || isNumber($code)) { |
||
231 | return $this->lexNumber($code, $line, $col, $prev); |
||
232 | } |
||
233 | |||
234 | // Name: [_A-Za-z][_0-9A-Za-z]* |
||
235 | if (isAlphaNumeric($code)) { |
||
236 | return $this->lexName($line, $col, $prev); |
||
237 | } |
||
238 | |||
239 | // Spread: ... |
||
240 | if ($this->bodyLength >= 3 && isSpread($this->body, $code, $this->pos)) { |
||
241 | return $this->lexSpread($line, $col, $prev); |
||
242 | } |
||
243 | |||
244 | // String: "([^"\\\u000A\u000D]|(\\(u[0-9a-fA-F]{4}|["\\/bfnrt])))*" |
||
245 | if (isString($this->body, $code, $this->pos)) { |
||
246 | return $this->lexString($line, $col, $prev); |
||
247 | } |
||
248 | |||
249 | // Block String: """("?"?(\\"""|\\(?!=""")|[^"\\]))*""" |
||
250 | if ($this->bodyLength >= 3 && isTripleQuote($this->body, $code, $this->pos)) { |
||
251 | return $this->lexBlockString($line, $col, $prev); |
||
252 | } |
||
253 | |||
254 | throw $this->createSyntaxErrorException(); |
||
255 | } |
||
256 | |||
257 | /** |
||
258 | * Creates an End Of File (EOF) token. |
||
259 | * |
||
260 | * @param int $line |
||
261 | * @param int $col |
||
262 | * @param Token $prev |
||
263 | * @return Token |
||
264 | */ |
||
265 | protected function createEndOfFileToken(int $line, int $col, Token $prev): Token |
||
266 | { |
||
267 | return new Token(TokenKindEnum::EOF, $this->bodyLength, $this->bodyLength, $line, $col, $prev); |
||
268 | } |
||
269 | |||
270 | /** |
||
271 | * Reads a punctuation token from the source file. |
||
272 | * |
||
273 | * @param int $code |
||
274 | * @param int $line |
||
275 | * @param int $col |
||
276 | * @param Token $prev |
||
277 | * @return Token |
||
278 | * @throws SyntaxErrorException |
||
279 | */ |
||
280 | protected function lexPunctuation(int $code, int $line, int $col, Token $prev): ?Token |
||
281 | { |
||
282 | switch ($code) { |
||
283 | case 33: // ! |
||
284 | return new Token(TokenKindEnum::BANG, $this->pos, $this->pos + 1, $line, $col, $prev); |
||
285 | case 36: // $ |
||
286 | return new Token(TokenKindEnum::DOLLAR, $this->pos, $this->pos + 1, $line, $col, $prev); |
||
287 | case 38: // & |
||
288 | return new Token(TokenKindEnum::AMP, $this->pos, $this->pos + 1, $line, $col, $prev); |
||
289 | case 58: // : |
||
290 | return new Token(TokenKindEnum::COLON, $this->pos, $this->pos + 1, $line, $col, $prev); |
||
291 | case 61: // = |
||
292 | return new Token(TokenKindEnum::EQUALS, $this->pos, $this->pos + 1, $line, $col, $prev); |
||
293 | case 64: // @ |
||
294 | return new Token(TokenKindEnum::AT, $this->pos, $this->pos + 1, $line, $col, $prev); |
||
295 | case 124: // | |
||
296 | return new Token(TokenKindEnum::PIPE, $this->pos, $this->pos + 1, $line, $col, $prev); |
||
297 | case 40: // ( |
||
298 | return new Token(TokenKindEnum::PAREN_L, $this->pos, $this->pos + 1, $line, $col, $prev); |
||
299 | case 41: // ) |
||
300 | return new Token(TokenKindEnum::PAREN_R, $this->pos, $this->pos + 1, $line, $col, $prev); |
||
301 | case 91: // [ |
||
302 | return new Token(TokenKindEnum::BRACKET_L, $this->pos, $this->pos + 1, $line, $col, $prev); |
||
303 | case 93: // ] |
||
304 | return new Token(TokenKindEnum::BRACKET_R, $this->pos, $this->pos + 1, $line, $col, $prev); |
||
305 | case 123: // { |
||
306 | return new Token(TokenKindEnum::BRACE_L, $this->pos, $this->pos + 1, $line, $col, $prev); |
||
307 | case 125: // } |
||
308 | return new Token(TokenKindEnum::BRACE_R, $this->pos, $this->pos + 1, $line, $col, $prev); |
||
309 | default: |
||
310 | throw $this->createSyntaxErrorException(); |
||
311 | } |
||
312 | } |
||
313 | |||
314 | /** |
||
315 | * Reads a name token from the source file. |
||
316 | * |
||
317 | * @param int $line |
||
318 | * @param int $col |
||
319 | * @param Token $prev |
||
320 | * @return Token |
||
321 | */ |
||
322 | protected function lexName(int $line, int $col, Token $prev): Token |
||
323 | { |
||
324 | $start = $this->pos; |
||
325 | $this->pos = $start + 1; |
||
326 | |||
327 | while ($this->pos !== $this->bodyLength && |
||
328 | ($code = charCodeAt($this->body, $this->pos)) !== null && |
||
329 | isAlphaNumeric($code)) { |
||
330 | ++$this->pos; |
||
331 | } |
||
332 | |||
333 | $value = sliceString($this->body, $start, $this->pos); |
||
334 | |||
335 | return new Token(TokenKindEnum::NAME, $start, $this->pos, $line, $col, $prev, $value); |
||
336 | } |
||
337 | |||
338 | /** |
||
339 | * Reads a number (int/float) token from the source file. |
||
340 | * |
||
341 | * @param int $code |
||
342 | * @param int $line |
||
343 | * @param int $col |
||
344 | * @param Token $prev |
||
345 | * @return Token |
||
346 | * @throws SyntaxErrorException |
||
347 | */ |
||
348 | protected function lexNumber(int $code, int $line, int $col, Token $prev): Token |
||
401 | ); |
||
402 | } |
||
403 | |||
404 | /** |
||
405 | * Reads digits from the source file. |
||
406 | * |
||
407 | * @param int $code |
||
408 | * @throws SyntaxErrorException |
||
409 | */ |
||
410 | protected function skipDigits(int $code): void |
||
411 | { |
||
412 | if (isNumber($code)) { |
||
413 | do { |
||
414 | $code = charCodeAt($this->body, ++$this->pos); |
||
415 | } while (isNumber($code)); |
||
416 | |||
417 | return; |
||
418 | } |
||
419 | |||
420 | throw $this->createSyntaxErrorException( |
||
421 | \sprintf('Invalid number, expected digit but got: %s.', printCharCode($code)) |
||
422 | ); |
||
423 | } |
||
424 | |||
425 | /** |
||
426 | * Reads a comment token from the source file. |
||
427 | * |
||
428 | * @param int $line |
||
429 | * @param int $col |
||
430 | * @param Token $prev |
||
431 | * @return Token |
||
432 | */ |
||
433 | protected function lexComment(int $line, int $col, Token $prev): Token |
||
434 | { |
||
435 | $start = $this->pos; |
||
436 | |||
437 | do { |
||
438 | $code = charCodeAt($this->body, ++$this->pos); |
||
439 | } while ($code !== null && ($code > 0x001f || $code === 0x0009)); // SourceCharacter but not LineTerminator |
||
440 | |||
441 | return new Token( |
||
442 | TokenKindEnum::COMMENT, |
||
443 | $start, |
||
444 | $this->pos, |
||
445 | $line, |
||
446 | $col, |
||
447 | $prev, |
||
448 | sliceString($this->body, $start + 1, $this->pos) |
||
449 | ); |
||
450 | } |
||
451 | |||
452 | /** |
||
453 | * Reads a spread token from the source. |
||
454 | * |
||
455 | * @param int $line |
||
456 | * @param int $col |
||
457 | * @param Token $prev |
||
458 | * @return Token |
||
459 | */ |
||
460 | protected function lexSpread(int $line, int $col, Token $prev): Token |
||
461 | { |
||
462 | return new Token(TokenKindEnum::SPREAD, $this->pos, $this->pos + 3, $line, $col, $prev); |
||
463 | } |
||
464 | |||
465 | /** |
||
466 | * @param int $line |
||
467 | * @param int $col |
||
468 | * @param Token $prev |
||
469 | * @return Token |
||
470 | * @throws SyntaxErrorException |
||
471 | */ |
||
472 | protected function lexString(int $line, int $col, Token $prev): Token |
||
473 | { |
||
474 | $start = $this->pos; |
||
475 | $chunkStart = ++$this->pos; |
||
476 | $value = ''; |
||
477 | |||
478 | while ($this->pos < $this->bodyLength && |
||
479 | ($code = charCodeAt($this->body, $this->pos)) !== null && !isLineTerminator($code)) { |
||
480 | // Closing Quote (") |
||
481 | if (34 === $code) { |
||
482 | $value .= sliceString($this->body, $chunkStart, $this->pos); |
||
483 | return new Token(TokenKindEnum::STRING, $start, $this->pos + 1, $line, $col, $prev, $value); |
||
484 | } |
||
485 | |||
486 | if (isSourceCharacter($code)) { |
||
487 | throw $this->createSyntaxErrorException( |
||
488 | \sprintf('Invalid character within String: %s.', printCharCode($code)) |
||
489 | ); |
||
490 | } |
||
491 | |||
492 | ++$this->pos; |
||
493 | |||
494 | if (92 === $code) { |
||
495 | // \ |
||
496 | $value .= sliceString($this->body, $chunkStart, $this->pos - 1); |
||
497 | |||
498 | $code = charCodeAt($this->body, $this->pos); |
||
499 | |||
500 | switch ($code) { |
||
501 | case 34: // " |
||
502 | $value .= '"'; |
||
503 | break; |
||
504 | case 47: // / |
||
505 | $value .= '/'; |
||
506 | break; |
||
507 | case 92: // \ |
||
508 | $value .= '\\'; |
||
509 | break; |
||
510 | case 98: // b |
||
511 | $value .= '\b'; |
||
512 | break; |
||
513 | case 102: // f |
||
514 | $value .= '\f'; |
||
515 | break; |
||
516 | case 110: // n |
||
517 | $value .= '\n'; |
||
518 | break; |
||
519 | case 114: // r |
||
520 | $value .= '\r'; |
||
521 | break; |
||
522 | case 116: // t |
||
523 | $value .= '\t'; |
||
524 | break; |
||
525 | case 117: // u |
||
526 | $unicodeString = sliceString($this->body, $this->pos + 1, $this->pos + 5); |
||
527 | |||
528 | if (!\preg_match('/[0-9A-Fa-f]{4}/', $unicodeString)) { |
||
529 | throw $this->createSyntaxErrorException( |
||
530 | \sprintf('Invalid character escape sequence: \\u%s.', $unicodeString) |
||
531 | ); |
||
532 | } |
||
533 | |||
534 | $value .= '\\u' . $unicodeString; |
||
535 | $this->pos += 4; |
||
536 | |||
537 | break; |
||
538 | default: |
||
539 | throw $this->createSyntaxErrorException( |
||
540 | \sprintf('Invalid character escape sequence: \\%s.', \chr($code)) |
||
541 | ); |
||
542 | } |
||
543 | |||
544 | ++$this->pos; |
||
545 | |||
546 | $chunkStart = $this->pos; |
||
547 | } |
||
548 | } |
||
549 | |||
550 | throw $this->createSyntaxErrorException('Unterminated string.'); |
||
551 | } |
||
552 | |||
553 | /** |
||
554 | * @param int $line |
||
555 | * @param int $col |
||
556 | * @param Token $prev |
||
557 | * @return Token |
||
558 | * @throws SyntaxErrorException |
||
559 | */ |
||
560 | protected function lexBlockString(int $line, int $col, Token $prev): Token |
||
598 | } |
||
599 | |||
600 | /** |
||
601 | * Creates a `SyntaxErrorException` for the current position in the source. |
||
602 | * |
||
603 | * @param null|string $description |
||
604 | * @return SyntaxErrorException |
||
605 | */ |
||
606 | protected function createSyntaxErrorException(?string $description = null): SyntaxErrorException |
||
607 | { |
||
608 | $code = charCodeAt($this->body, $this->pos); |
||
609 | return new SyntaxErrorException( |
||
610 | $this->source, |
||
611 | $this->pos, |
||
612 | $description ?? $this->unexpectedCharacterMessage($code) |
||
613 | ); |
||
614 | } |
||
615 | |||
616 | /** |
||
617 | * Report a message that an unexpected character was encountered. |
||
618 | * |
||
619 | * @param int $code |
||
620 | * @return string |
||
621 | */ |
||
622 | protected function unexpectedCharacterMessage(int $code): string |
||
623 | { |
||
624 | if (isSourceCharacter($code) && !isLineTerminator($code)) { |
||
625 | return \sprintf('Cannot contain the invalid character %s.', printCharCode($code)); |
||
626 | } |
||
627 | |||
628 | if ($code === 39) { |
||
629 | // ' |
||
630 | return 'Unexpected single quote character (\'), did you mean to use a double quote (")?'; |
||
631 | } |
||
632 | |||
633 | return \sprintf('Cannot parse the unexpected character %s.', printCharCode($code)); |
||
634 | } |
||
635 | |||
636 | /** |
||
637 | * |
||
638 | */ |
||
639 | protected function skipWhitespace(): void |
||
664 | } |
||
665 | } |
||
666 | } |
||
667 | } |
||
668 |